分子生物学

RNA detection techniques have expanded into a diverse methodological landscape spanning hybridization, amplification, imaging, and sequencing. In this review, we provide a method‑centric synthesis of the major technologies that define this landscape, emphasizing how each method’s core principle, practical strengths, and sources of variability shape its reproducibility. Beginning with foundational approaches, we trace the development of isothermal amplification, quantitative and digital PCR, microarrays, single‑molecule imaging, multiplexed spatial methods, and amplification‑free digital quantification. We then examine the transformative impact of bulk, single‑cell, long‑read, direct‑RNA, and spatial transcriptomics, as well as CRISPR‑based detection and metabolic labeling for RNA dynamics. Across these technologies, we focus on reproducibility as a defining dimension of evaluation: mature methods benefit from established standards, whereas newer approaches remain pre‑standardization and require careful, experiment‑specific controls. Rigorous method selection must be guided by the biological question, required resolution, sample constraints, and the maturity of each method’s reproducibility framework. We conclude that RNA detection methods form interconnected methodological paths of problem‑solving rather than simple replacements.

The PrimeFlow^TM assay is a flow cytometry–based method for the co-detection of RNAs and proteins in cells. When combined with cell characterization by immunophenotyping, PrimeFlow^TM can be used to simultaneously detect RNA and proteins in a cell type–specific manner in complex heterogeneous samples, offering an advantage over bulk tissue analysis methods. Here, we describe the implementation of the PrimeFlow^TM assay protocol for the detection of transgene mRNA and protein expression in spleen samples from mice treated in vivo with luciferase mRNA-lipid nanoparticles (LNPs). This protocol involves spleen tissue dissociation for cell isolation, followed by cell fixation and permeabilization to allow immunolabeling of intracellular luciferase protein. The immunophenotyping strategy is based on immunolabeling with mouse CD marker antibodies for the identification of T cells, B cells, monocytes, granulocytes/macrophages, NK cells, and non-hematopoietic cells. The RNAs of luciferase and a housekeeping gene, β-actin, are detected with sequence-specific probe sets by employing sequential oligonucleotide annealing steps and fluorescent labeling using a branched DNA (bDNA) technology. Samples are analyzed by flow cytometry. Based on our analysis, we conclude it is feasible to apply the PrimeFlow^TM approach for evaluating successful drug targeting to the cell types of interest and any potential differences in the kinetics of RNA delivery and protein expression in various tissue cells, supporting the discovery and development of RNA therapeutics.

RNA alternative splicing (AS) is an essential process that expands transcriptomic and proteomic diversity in eukaryotic cells and contributes to cellular heterogeneity across physiological and pathological conditions in humans. With the advent of single-cell RNA sequencing (scRNA-seq), it has become possible to study AS at cellular resolution, although robust and standardized analytical workflows remain to be developed. Here, we present a stepwise protocol for analyzing AS in single cells from pediatric high-grade gliomas (pHGGs) harboring the histone H3.3 lysine 27-to-methionine (H3.3K27M) mutation using SMART-Seq2 scRNA-seq data. Starting from raw sequencing reads, the workflow includes read alignment, gene-level quantification, splice junction and intron quantification, and single-nucleotide variant-based mutation detection. Gene expression–based clustering and cell-type annotation are performed by using the Seurat R package. AS analysis in tumor cells is then conducted using the MARVEL R package in combination with customized scripts to calculate percent spliced-in (PSI) values, identify variable AS events, perform dimensionality reduction, cluster cells, conduct differential AS analysis, and visualize splicing patterns. This protocol provides a reproducible and comprehensive framework for dissecting AS dynamics at single-cell resolution. It is readily adaptable to other SMART-Seq2 datasets and facilitates systematic investigation of splicing heterogeneity in diverse biological contexts.

Pathogen gene expression from host samples is often challenging to study due to low signal and high host RNA background. PCR probes have been recently used to hybridize and extract bacterial sequences from next-generation sequencing (NGS) libraries generated from in vitro and animal models of infection; however, these strategies require purchasing commercially synthesized probes that often do not capture the entire transcriptome. Transcript-capture sequencing is a novel capture approach for extracting RNA of a target bacterial species from samples in which there is substantial contamination by the host or other microbes. Biotinylated 150-base-pair DNA probes are generated in-house from bacterial DNA spanning the entire bacterial genome. Probes are hybridized to the cDNA of NGS sequencing libraries prepared from host samples to capture and enrich for bacterial-specific RNA reads before sequencing. This method results in a >200-fold increase in bacterial RNA reads from infected host samples (including in vitro, animal, and human samples) and generates complete bacterial transcriptomes with high gene coverage (>80%). Use of this protocol on infected host samples reveals a snapshot of bacterial activity during disease that may improve understanding of the physiological state of pathogens within their hosts.

Transfer RNAs (tRNAs) are important regulators of translation and cellular function. Several high-throughput sequencing methods have been developed to quantitatively analyze tRNA isoacceptors in cells. However, the strong secondary structures and extensive post-transcriptional modification of most tRNA molecules present significant challenges for many reverse transcriptases, negatively impacting sequencing library preparation and causing quantification biases. Currently, the field utilizes processive next-generation reverse transcriptases (ngRTs), such as Induro (New England Biolabs) and UltraMarathonRT (RNAConnect), to address these issues. Despite being used in multiple protocols, these commercial products face little competition and remain costly. However, non-commercial alternatives, such as the original MarathonRT (MRT), are available from gene repositories. MRT is a next-generation reverse transcriptase derived from the Eubacterium rectale group II intron maturase, which can read through RNA secondary structures and chemical modifications. Here, we present a simplified expression and purification protocol for producing highly active MRT that is stable over 1 year. This cost-effective protocol yields a heterogeneous protein preparation with no discernible competing enzymatic activities; it mitigates previously reported precipitation issues, saving one day of laboratory work and eliminating two chromatography-based purification steps. Moreover, the use of the resulting protein preparation has been verified in the mim-tRNAseq pipeline, where it was shown to perform equally to the commercial alternatives Induro and UltraMarathonRT. In addition, we have developed a simple and cost-effective assay for measuring the enzymatic activity of MRT, allowing for batch comparison.

Fluorescence in situ hybridization (FISH) can be employed to study the expression and subcellular localization of nucleic acids by using labeled antisense strands that hybridize with the target RNA or DNA molecules. Likewise, immunofluorescence antibody staining (IF) takes advantage of the specific interaction between a fluorophore-labeled antibody and its corresponding antigen. This protocol reports the combination of RNA-FISH and IF antibody staining for simultaneous detection of both RNA transcripts and proteins of interest in routine formalin-fixed paraffin-embedded (FFPE) bone marrow biopsy samples. Herein, we provide a detailed description of the methodology that we have developed and optimized to study the spatial expression of two transcripts—TGFB1 and PDGFA1—in human hematopoietic (CD45⁺) and non-hematopoietic (CD271⁺) cells in the bone marrow of patients with acute lymphoblastic leukemia (ALL).

Understanding gene expression within defined neuronal populations is essential for dissecting the cellular and molecular diversity of the brain. mRNA assays provide a direct readout of gene expression, capturing transcriptional changes that may precede or occur independently of protein abundance, whereas protein assays reflect the cumulative effects of translation, modification, and degradation. Moreover, in histological analysis, immunohistochemical protein detection results in visually diffuse labeling, which makes it difficult to quantitatively assess levels and locations of expression at high resolution. Here, we present a protocol that allows for mRNA detection in single neuronal cell types with a high degree of sensitivity and anatomical resolution. This protocol combines fluorescent in situ hybridization (FISH) with immunohistochemistry (IHC) on the same tissue section. Briefly, FISH is carried out by ACDBio RNAscope^® fluorescent in situ hybridization technology, which involves processing the tissue sections, followed by signal amplification. This involves target retrieval, probe hybridization, and signal enhancement. Then, the tissue section is processed for IHC, which involves blocking nonspecific sites and incubation with primary antibodies, followed by development of a fluorescent signal with secondary antibodies. Typically, visual mRNA detection with FISH can be seen as individual puncta, whereas targeting the protein with an antibody results in filled cells or processes. The variation in staining pattern allows for the quantification of distinct mRNA transcripts within different neuronal populations, which renders co-localization analyses easy and efficient.

Evaluating RNA–protein interactions is key to understanding post-transcriptional gene regulation. Electrophoretic mobility shift assays (EMSAs) remain a widely used technique to study these interactions, revealing information about binding affinities and binding modalities, including cooperativity and complex formation. Here, we detail, in a step-by-step protocol, how to perform EMSAs. We describe how to generate, purify, and quantitate ³²P-radiolabeled RNA by in vitro transcription, as well as the expression and purification of recombinant RNA-binding proteins in E. coli using ELAV as an example. We then describe how to set up binding reactions using serial dilutions in a microtiter plate format of recombinant ELAV and in vitro–transcribed RNA and how to perform EMSAs using native low-crosslinked acrylamide gels, with detailed graphically supported instructions and troubleshooting guides.

The genomes of RNA viruses can fold into dynamic structures that regulate their own infection and immune evasion processes. Proximity ligation methods (e.g., SPLASH) enable genome-wide interaction mapping but lack specificity when dealing with low-abundance targets in complex samples. Here, we describe HiCapR, a protocol integrating in vivo psoralen crosslinking, RNA fragmentation, proximity ligation, and hybridization capture to specifically enrich viral RNA–RNA interactions. Captured libraries are sequenced, and chimeric reads are analyzed via a customized computational pipeline to generate constrained secondary structures. HiCapR generates high-resolution RNA interaction maps for viral genomes. We applied it to resolve the in vivo structure of the complete HIV-1 RNA genome, identifying functional domains, homodimers, and long-range interactions. The protocol's robustness has been previously validated on the SARS-CoV-2 genome. HiCapR combines proximity ligation with targeted enrichment, providing an efficient and specific tool for studying RNA architecture in viruses, with broad applications in virology and antiviral development.

Functional enrichment analysis is essential for understanding the biological significance of differentially expressed genes. Commonly used tools such as g:Profiler, DAVID, and GOrilla are effective when applied to well-annotated model organisms. However, for non-model organisms, particularly for bacteria and other microorganisms, curated functional annotations are often scarce. In such cases, researchers often rely on homology-based approaches, using tools like BLAST to transfer annotations from closely related species. Although this strategy can yield some insights, it often introduces annotation errors and overlooks unique species-specific functions. To address this limitation, we present a user-friendly and adaptable method for creating custom annotation R packages using genomic data retrieved from NCBI. These packages can be directly imported as libraries into the R environment and are compatible with the clusterProfiler package, enabling effective gene ontology and pathway enrichment analysis. We demonstrate this approach by constructing an R annotation package for Mycobacterium tuberculosis H37Rv, as an example. The annotation package is then utilized to analyze differentially expressed genes from a subset of RNA-seq dataset (GSE292409), which investigates the transcriptional response of M. tuberculosis H37Rv to rifampicin treatment. The chosen dataset includes six samples, with three serving as untreated controls and three exposed to rifampicin for 1 h. Further, enrichment analysis was performed on genes to demonstrate changes in response to the treatment. This workflow provides a reliable and scalable solution for functional enrichment analysis in organisms with limited annotation resources. It also enhances the accuracy and biological relevance of gene expression interpretation in microbial genomics research.